MEMS free-space optical switches can be categorized into two major branches: the planar matrix (2-dimensional) approach, and the beam-steering (3-dimensional) approach. The 2D approach typically involves mirrors that move between on and off position, while the 3-D approach typically involves mirrors that tilt over a continuous range of angles to deflect optical signals from one fiber array to another. The 3-D approach relies on accurate control of mirror position to minimize optical loss from the coupling of photons from one fiber to another.
Fiber optic communications systems are subject to faults that interrupt signal traffic. The fault may occur in the optical switch or in some other part of the system. In both switching approaches it is useful for, fault detection purposes, to know whether a given mirror actuating mechanism has failed. One way to determine this is to sense the position of the mirror to determine whether it is in a desired state. If the mirror is not in the desired state, a fault in the mirror mechanism may be determined and signal traffic may be routed around the faulty mirror.
Most of these MEMS optical elements have used some variation of sensing capacitance or piezoresistance as a means of detecting the angular position of the optical element. In the 2D approach, to perform accurate capacitance sensing the signal lines have to be shielded which adds significantly to the complexity of the MEMs die. Second, the capacitive sensing is highly non linear and the sensitivity degrades significantly at large angular deviations from the ideal final position. The piezoresistive sensors have smaller signal gain making them susceptible to noise and cross-talk.
Thus, there is a need in the art, for a new method and apparatus for sensing the angular position of a MEMS optical element and an optical switch incorporating same.